Micron‐Textured Ambipolar Photodetectors Enabling ∼20 µs Photonic Adaptation
ABSTRACT The rapid development of artificial intelligence calls for compact, flexible, and intelligent photodetectors. However, strong background illumination can overwhelm weak optical signals, while downstream compensation imposes substantial computational burdens, thereby introducing considerable processing latency. Here, we demonstrate an ultrathin (8 µm), mechanically flexible ambipolar electronic vision (AEV) system that suppresses background optical interference directly at the photodetection front end. Symmetric back‐to‐back Schottky barriers minimize the net vertical internal electric field, while a self‐assembled donor‐enriched micron‐scale texture generates a small surface‐potential gradient (∼10 mV) to balance carrier transport and collection. This mechanism stabilizes illumination‐direction‐dependent photocurrent polarity near zero bias, reduces the minimum irradiance required for ambipolar operation from 175 mW cm −2 to 0.1 mW cm −2 , corresponding to an approximately 1750‐fold reduction. Through the device‐level superposition of oppositely signed photocurrents, the ultrathin, flexible AEV array directly suppresses background‐induced photocurrent offsets and exhibits bidirectional response times of ∼20 µs, enabling rapid photonic adaptation. The ambipolar response remains stable after array integration and 3500 bending cycles. Nearly 100% recognition accuracy is maintained under strong background illumination of ∼11 mW cm −2 , establishing a device‐level strategy for rapid and reliable visual perception in flexible and conformable electronic vision systems.
Authors
- Xiaosheng Fang (ORCID: https://orcid.org/0000-0003-3387-4532)
- Ziqing Li (ORCID: https://orcid.org/0000-0002-8126-2728)
- Gang Wang (ORCID: https://orcid.org/0000-0002-0833-4887)
- Limin Wu (ORCID: https://orcid.org/0000-0001-8495-8627)
- Ming Deng
- Tingting Yan
- Yarong Gu
- Bobo Li
- Junhao Chu
Institutions
- Fudan University (CN)
- Inner Mongolia University (CN)
- Wuhan University (CN)
Publication Details
- Journal
- Advanced Materials
- Published
- 2026-09-25
- DOI
- https://doi.org/10.1002/adma.75139
- Primary Topic
- Neural Networks and Reservoir Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00